Date: Sun, 20 Jun 1999 16:34:24 -0700 (PDT) From: Tani Hosokawa <unknown@riverstyx.net> To: Jonathon Doran <doranj@Colorado.EDU> Cc: questions@freebsd.org Subject: Re: Nt outperforms Linux for web/file server? Message-ID: <Pine.LNX.4.10.9906201608080.1606-100000@avarice.riverstyx.net> In-Reply-To: <199906202303.RAA11752@ucsu.Colorado.EDU>
next in thread | previous in thread | raw e-mail | index | archive | help
Apologies if this was meant to be a private e-mail... it didn't look like it. Now, for my soapbox... anyone seen my soapbox? Ah, there it is... pass it over. Apache's slower than other webservers. I can say this because a multitude of benchmarks support my case. Even the Apache developers support this. In the discussions that took place on linux-kernel and new-httpd people said that Apache is not a high performance webserver. In the Apache Performance Tuning FAQ it says that Apache is not a high performance web server. As for Linux at that point in time being buggy scalability-wise (pre-2.2.7), that's also documented in the various mailing lists. http://www.acme.com/software/thttpd/benchmarks.html That is a set of benchmarks that show that Apache is not a high performance web server, thoroughly beaten down by thttpd, Boa, mathopd, and Zeus. From the Apache performance tuning FAQ (http://www.apache.org/docs/misc/perf-tuning.html): "Apache is a general webserver, which is designed to be correct first, and fast second. " From Dean Gaudet, one of the Apache core developers, on linux-kernel (Fri, 16 Apr 1999 19:28:35 PDT): '> OK. Most of the important points have been covered already. Especially > the tuning of the apache server itself is one of the most significant > issues. Uh I dunno. Unless by tuning you mean "replace apache with something that's actually fast" ;)' Point being, Apache's not fast. It's pretty fast, and it's stable, and it's easy to program for -- hence why I use it to serve 80m hits/day (just about saturating a 100Mb connection) on a couple machines, instead of jumping into thttpd and hacking code to get the same performance out of one server. Which I have done when I felt it prudent, before someone starts criticizing my "lack of hacker work ethic" or some other such bullshit. I hate it when people deny the obvious just because it's bad ol' (insert your favorite "bad guy" here, in this case Microsoft). It's naive and detrimental to everyone involved. I have yet to see Apache serve 1000 req/sec, and it probably won't inside the next 6 months. That's not me saying "Apache sux", it's me being realistic. This isn't me condemning the use of Apache either. There are plenty of tools available to the world. Find the tool that suits your purpose and use it. But don't fall into the trap of only using that one tool for everything. On Sun, 20 Jun 1999, Jonathon Doran wrote: > Neither of you guys are making any sense... > > Tani says > "Apache Sux" or something similar without giving any details, or > attempting to back up this claim. (Ad Homimum) > > Micheal says: > "Everyone is using it, so its great", an argument which could be > applied to cigarettes or disco in the 70's. (Bandwagon) > > I doubt most people care how fast Apache is, since they don't serve > a high enough volume of pages, or their connection saturates before Apache > does. > > Those who do care, have the sources. And I trust they'll patch it. > > And from my point of view, it works well for my application. So I'm happy > with it. > > Jon Doran --- tani hosokawa river styx internet To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.LNX.4.10.9906201608080.1606-100000>